Scientific Advisory Committee on Medical Devices used in Cardiovascular Systems - Record of Proceedings - November 28, 2014

Committee Members Present: John Ducas (Chair), Renzo Cecere, Christopher Feindel, Alan Menkis, Brent Mitchell, Barry Rubin, Raymond Yee, John Webb
Invited Guests: Thomas Forbes, Reda Ibrahim, Prasad Jetty, Raj Padwal
Health Canada Representatives:
- Office of Science: Hripsime Shahbazian
- Office of Planning, Performance, and Review Services: Caroline Hunt
- Medical Devices Bureau: Carey Agnew,Kevin Day,Ian Aldous, Patrick Fandja, Ian Glasgow, Ian Grimwood, Jianming Hao, Karen Kennedy, Mark Korchinski, Christine Lefebvre, James McGarrity, Catherine Milley, Michael Rosu-Myles, Philip Neufeld, Jason Pearman, Maurice Sylvain, KokSwang Tan, Lanyi Xu, Brian Wong
Regrets: Luc Bilodeau, Eric Cohen, Marino Labinaz, Joaquim Miró

Abbreviations used in this record:

AAA:: abdominal aortic aneurysm
AMI:: acute myocardial infarction
ATP:: anti-tachycardia pacing
AVR:: aortic valve replacement
BMC:: bone marrow cell
CABG:: coronary artery bypass grafting
CACs:: circulating angiogenic cells
CADTH:: Canadian Agency for Drugs and Technologies in Health
CPC:: cardiac progenitor cell
CSF:: cerebrospinal fluid
CT:: computed tomography
DEB:: drug eluting balloons
ECC:: extracorporeal circulation
EEC:: endovascular elephant trunk completion
EPC:: endothelial progenitor cell
EVAR:: evolution of endovascular aneurysm repair
HF:: human factors
ICD:: implantable cardioverter-defribrillator
IFU:: instructions for use
LVEF:: left ventricular ejection fraction
MDB:: medical devices bureau
MI:: myocardial infarction
MEP:: motor-evoked potentials
MRI:: magnetic resonance imaging
MSC:: mesenchymal stem cell
OEC:: open elephant trunk completion
OPC:: objective performance criteria
RCT:: randomized clinical trial
SAC-MDUCS:: scientific advisory committee on medical devices used in the cardiovascular system
SAP:: special access program
SCI:: spinal cord fluid
SEP:: somatosensory evoked potentials
SVT:: supraventricular tachycardia
TAA:: thoracic aortic aneurysm
TAVI:: transcatheter aortic valve implantation
VF:: ventricular fibrillation
VT:: ventricular tachycardia
VTA:: ventricular tacky arrhythmia

1. Opening Remarks & Welcome

Dr. John Patrick Stewart, A/DG, TPD

The Interim Senior Executive Director of the Therapeutic Products Directorate (TPD), Dr. John Patrick Stewart opened the meeting and welcomed the committee members and guest speakers. He thanked them for sharing their time and expertise and acknowledged the efforts all participants had made to attend the meeting. Dr. Stewart highlighted the importance of the Scientific Advisory Committee on Medical Devices Used in the Cardiovascular System (SAC-MDUCS) to Health Canada’s regulatory review and decision-making process. He provided the Committee with a brief update on the Government’s current position as well as next steps regarding Bill C-17, a piece of legislation created to enhance the safety of Canadians by expanding Government authority over the product lifecycle of a drug. He noted that Health Canada will continue to seek committee advice to plan future approaches to address device related issues. Dr. Stewart concluded by wishing the committee members and guest speakers fruitful deliberations.

2. Review of the Agenda, Affiliations and Interests Declarations, and Confidentiality Agreement

Dr. John Ducas, Committee Chair

The Chair reviewed the agenda items with the committee. The agenda was accepted as established.

Verbal affiliations and interest declarations were completed. No conflicts were identified at this time. Members were also asked to disclose any conflicts that may arise as the meeting proceeds.

Presentations are available upon request.

The responses outlined below are representative of the discussions/recommendations generated by the committee.

3. Summary of How Health Canada Has Used Information Generated from Previous SAC-MDUCS Meetings

Kevin Day, Medical Devices Bureau (MDB)

Mr. Kevin Day presented a brief summary of how previous advice and recommendations provided by the SAC-MDUCS have been considered by Health Canada.

The following topics were noted:

DF-4 Connectors: previous discussions and recommendations regarding requirements for clinical evidence and post-market follow-up for DF-4 connectors were considered in license rejection decision after manufacturers failed to provide post-market follow-up.
Drug Eluting Balloons: drug eluting balloons (DEBs) have been a very important area in discussions with the manufacturers and feedback received from the committee regarding enforcing superiority against angioplasty and minimum trial sizes, with one year follow-up have had implications on clinical trial designs for licensing purposes.
Suturless Valves: discussions on suturless valves compared to surgical valves, compared to TAVI and the importance of longer term follow-up data with evidence of morbidity and mortality associated with these devices have provided a better understanding of this technology before issuing a license.
Stem Cells: information obtained from previous discussions on stem cells has been very helpful regarding a recent clinical trial in which stem cells were being generated by a medical device.
AAA Devices: currently using discussions and recommendations from the committee to aid in reviews of additional branch devices.
Implantable cardioverter-defibrillators (ICDs): recommendations received from the committee regarding ICDs have been helpful in determining appropriate patient population.

The Medical Devices Bureau (MDB) is using the recommendations and advice provided by the SAC-MDUCS when discussing clinical trials or licensing requirements with Industry. The Committee’s input also ensures that good clinical oversight is continually being applied towards decision making processes.

Mr. Day thanked members for their ongoing dedication to the committee.

4. Renal Denervation - Next Steps for a Regulatory Body, Typical Use, Requirements for New Licenses

Dr. Raj Padwal, Guest Speaker

The first renal denervation system was licensed in Canada in 2012 following clinical data from three studies including one randomized unblinded study and a discussion on this topic by this Scientific Advisory Committee.

An overview of Health Canada’s approach to licensing renal denervation technologies was presented.

In 2014, additional data from a larger single blinded study put the clinical effectiveness of renal denervation into question. In light of recent developments, the committee has been asked to address several questions posed by Health Canada.

Dr. Raj Padwal was invited as a guest speaker to address this topic. He thanked the committee for the invitation and provided a short bio. After declaring his affiliations Dr. Padwal proceeded to address the questions posed by Health Canada.

Dr. Padwal provided a brief background information to help frame the topic of his presentation. He noted that:

Surgical sympathectomy was one of the earliest treatments for high blood pressure, predating drug therapy, and was used for malignant hypertension (1930). This was a precursor for renal intervention.
Renal denervation is a minimally invasive procedure that disrupts both the afferent and efferent nerves, as they are both implicated in the pathway of high blood pressure.
- Currently, there are no tests to verify whether denervation has been successful.
Resistant hypertension is the failure to achieve blood pressure target (<140/90mm Hg) despite treatment with three antihypertensive drugs (including a diuretic) at optimal doses.
- Hypertension affects 20% of Canadian adults.
- When adherence, lack of optimal therapy, and white coat effect are taken into account, the number of adult Canadians with true resistant hypertension would likely be around 0.2%.
Two main factors that increase susceptibility to resistant hypertension are:
- Age: geriatric patients have stiff vessels, which have been correlated with high blood pressure.
- Weight: obese patients (usually young) are at risk of resistant hypertension since the path of physiology involves central adiposity.

Dr. Padwal proceeded to address the topics presented by Health Canada.

Provide and discuss recommendations for clinical data requirements for licensing for new renal denervation systems (e.g., non-inferiority compared to a currently licensed renal denervation system, superiority compared to a sham control, etc.)

Dr. Padwal provided the following recommendations regarding clinical data requirements and trial design for these new systems:

Trials should be double blinded;
Comparators should be optimal medical therapy;
Study populations should include patients with resistant hypertension and these individuals should have a 24-hour ambulatory blood pressure (to rule out white coat effect);
Adherence should be assessed and non-adherent patients excluded;
Blood pressure measurements should be carried out with careful attention;
Interventionalist should be as experienced as possible.

It was noted that the SYMPLICITY-3 methodology should be considered Industry standard. Also, non-inferiority should play no role in clinical data requirements for new renal denervation systems.

What endpoints should be studied in clinical trials for hypertension and what represents a clinically significant improvement (e.g., office SBP, ABPM)? What are the incremental benefits of lowering the different measures of blood pressure? Will these incremental benefits be observed consistently across different groups of hypertensive patients?

Dr. Padwal noted that systolic blood pressure is considered the strongest indicator of outcome, therefore, with regards to endpoints in clinical trials for hypertension, systolic blood pressure ascertained through ambulatory blood pressure monitoring (ABPM) could be an acceptable primary endpoint. Automated office measurements should not be considered an acceptable primary endpoint since the white coat effect introduces subjectivity. It could, however, be considered a secondary endpoint.

Hypertension studies should assess/provide:

Morbidity and mortality (safety);
Short-term data (complications);
Long-term data (renal function, renal artery stenosis).

The use of “support of blood pressure” data is a good validated surrogate endpoint. The absolute increased risk when blood pressure increases is much greater in the elderly than it is in the young in terms of stroke mortality events.

Dr. Padwal noted that individuals with a high 5 or 10-year risk of cardiovascular disease will avoid the highest number of events through blood pressure reduction. Therefore, the best patient population for hypertension studies would be high-risk patients.

Discuss the risks and benefits of renal denervation given available clinical information. Provide insights and interpretations of difference between HTN-1 and -2 compared to SYMPLICITY HTN-3. What is your understanding on why the positive results from HTN-1 and HTN-2 were not seen in HTN-3?

Dr. Padwal explained that these differences in risks and benefits between renal denervation studies largely relate to methodology. He further noted that the operator experience plays a significant role (e.g., many operators in the SYMPLICITY HTN-3 had performed less than two procedures). It was recommended that interventionalists with significant experience in the renal arteries carry out the procedure.

Dr. Padwal noted that currently there is no test available allowing assessment of proper and effective execution of the renal denervation procedure. As a result, benefits of the procedure remain unclear. Despite an unknown level of efficacy, however, the safety of the procedure has been assessed and the risks appear low. The major concern has been deterioration in renal function. The caveat is that long-term data for large patient numbers are not available.

Dr. Padwal provided a list of potential contributing factors to differences between studies:

No assessment test available;
Relying on non-standardized blood pressure measurements;
White coat effect and non-adherence;
- SYMPLICITY HTN-3 had two screening visits and implemented a 24 hour ambulatory measurement as the second screening to limit the number of patients with white coat effect and non-adherence.
- Patients experiencing white coat effect will not present with elevated blood pressure on an ambulatory measure.
Improper training pathways prior to implementing the procedure.
The SYMPLICITY HTN-2 trial had follow-up assessors who were not blinded.

The following optimal regimen was recommended:

Assess for non-adherence;
Perform 24 hour ABPM;
Discontinue interfering drugs;
Rule out secondary causes;
Optimize health behaviors;
Use long acting synergistic combinations (ACE or ARB/DHP CCP/diuretic);
Simplify dosing schedule;
Nocturnal dosing;
Minimize cost.

Are there specific shortcomings in the design of currently available renal denervation systems that might significantly impact their safety and effectiveness? Please discuss any new designs in development that might be promising.

Dr. Padwal recommended that an assessment program/measurement must be developed to determine the effectiveness of the procedure. The suggestion of using a urine metabolite to determine whether denervation has taken place was provided.

It was also recommended that in addition to developing an assessment program/measurement, there should be more experimental animal model data obtained.

Currently, what patient populations should be treated with renal denervation in Canada? What qualifies a patient as having refractory hypertension in terms of blood pressure levels and medical therapy treatment?

Dr. Padwal suggested that patients who refuse other treatment (low adherence to pharmaceutical treatment) could be considered appropriate candidates for renal denervation as a last resort.

He explained that the level of pathogenesis of a given patient with hypertension related to the sympathetic nervous system remains unknown. A number of studies have demonstrated that the sympathetic nervous system is implicated, however, other contributing factors are not accounted for (e.g., it is unknown what percentage a patient’s hypertension is caused by the sympathetic nervous system).

It was noted that no patient population clearly stands out because the efficacy is uncertain. [The SYMPLICITY criteria of SBP >160 (150 in DM) and refractor HTN is reasonable to study as long as truly refractory]. Thus, the procedure should be performed mostly within the confines of research studies. Because of concerns regarding efficacy, it is not recommended that all patients with refractory hypertension be considered for RDN.

It was suggested that a case could be made for allowing the procedure to be performed in patients with mod-severe elevations in BP (160/100 mm Hg or higher; especially if target organ damage is present) and who are clearly not able to tolerate or adhere to therapy. Therefore, despite questions regarding efficacy, a case could be made that the benefits outweigh the risks even if the procedure largely produces some kind of ‘placebo’ response. But, few patients would be in this category.

Provide and discuss recommendations for action to be taken for currently licensed renal denervation systems in light of the currently available clinical trial data.

It was suggested that since these devices are licensed, one could allow these to remain licensed to allow ‘compassionate’ or ‘last resort’ cases to be done. These should only be done in the context of:

A fully informed patient;
A specialty hypertension clinic to manage the patient;
Experienced interventionalists.

Do you feel Health Canada should require more rigorous clinical trial data from novel cardiovascular medical device technology prior to licensing them for use in Canada?

It was recommended that the SYMPLICITY-HTN-3 trial be used as the standard with the addition of the requirement for an experienced operator.

Dr. Padwal concluded by acknowledging that truly resistant hypertension is not as common as believed. He noted that better research data for renal denervation is needed (both mechanistic and clinical) and that further research, such as developing an appropriate procedure-assessment tool, is required before routine clinical use can be implemented. Finally, more attention should be paid to proper methodology in clinical studies.

Dr. Padwal left the meeting.

5. Use of LAA Closure Devices, Clinical Requirements and Patient Populations

Dr. Reda Ibrahim, Guest Speaker

What level of clinical evidence is required to be confident in the safety and effectiveness of implantable endocardial Left Atrial Appendage (LAA) closure devices? The target population for these devices is typically patients with non-valvular atrial fibrillation who are at increased risk of stroke due to thrombotic emboli originating from the LAA. The gold standard for prevention of strokes in the non-valvular AF patients at greatest risk of stroke according to their CHADS₂/CHA₂DS₂-Vasc score is medical anticoagulation or antiplatelet (e.g., aspirin, warfarin).

Dr. Ibrahim thanked the committee for inviting him as a guest speaker. Before proceeding to address the topics provided by Health Canada he provided a short bio and declared his affiliations.

Dr. Ibrahim provided a brief overview of anticoagulation in atrial fibrillation (AF) as well as the limitations of the anticoagulant treatment and he outlined the rationale for Left Atrial Appendage Occlusion (LAAO) as a response to AF being associated with poor atrial contractility. He described type of patients who are at high risk of bleeding.

As a clinician, what level of evidence would you like to see to be confident in the safety and effectiveness of the implantable LAA closure devices? Please comment on:

Appropriate clinical endpoints (e.g., ischemic or hemorrhagic stroke);

Dr. Ibrahim noted that regarding the effectiveness of the implantable LAA closure devices, the following clinical endpoints should be required:
- Evidence of stroke prevention (ischemic and hemorrhagic);
- Evidence of systemic embolism prevention;
- Evidence of death prevention (total and CV).
In addition, regarding safety of the implantable LAA closure devices he recommended that the following primary endpoints should be required:
- Evidence of minimal complications per procedure;
- Evidence of minimal bleeding episodes.
Non-inferiority or superiority trial designs;

Dr. Ibrahim suggested that a superiority trial run against a placebo or an antiplatelet agent may be considered ethical; however, if the device is compared to warfarin or novel oral anticoagulants (NOACs), a non-inferiority trial would be sufficient.
Appropriate comparators for clinical safety (warfarin, NOACs);

He noted that NOACs are appropriate comparators for clinical safety as they are a safer choice when decreasing intracardial bleeding.

Dr. Ibrahim recommended Apixaban as the NOAC of choice.
Appropriate patient population;

Dr. Ibrahim suggested that an appropriate patient population would include patients with an elevated risk of bleeding (e.g., geriatric patients).

He noted that antiplatelet agents are not sufficient on their own to protect patients against stroke.
Appropriate duration of follow-up (both for pre-market and post-market monitoring). It was suggested that the appropriate duration of pre-market follow-up should be at least 6 weeks (preferably 1 year). Most complications related to LAA closure occur within 6 weeks (peri-procedural) and most of the patients have a TEE at 45 days to exclude thrombus and/or significant residual flow around the device.

It was suggested that within the field of LAA closure, post-approval studies should be a condition to pre-market approval to assess:
- Device safety;
- Effectiveness;
- Reliability in the real-world setting;
- The learning curve;
- Effectiveness of training programs;
- How well device performs in certain groups of patients.
It was noted that the appropriate duration of post-market follow-up should be at least 1 year (preferably 5 years).

Are there patients that can be appropriately treated today with some of the devices in development that have sufficient clinical evidence to be licensed in Canada? Discuss the available clinical evidence that provides you with sufficient confidence to use these devices.

Dr. Ibrahim commented on numerous sources of clinical evidence including:

Prevention in patients with non-valvular atrial fibrillation (PREVAIL) trial;
Aspirin and Plavix^® registry (ASAP registry);
Multicenter experience with the amplatzer cardiac plug (ACP);
Canadian registry.

He discussed the Randomized Trial of LAA Closure vs. Warfarin for Stroke/Thromboembolic Prevention in Patients with Non-valvular Atrial Fibrillation (PREVAIL). The PREVAIL trial was a prospective randomized evaluation of the WATCHMAN™ LAA Closure Device in patients with atrial fibrillation versus long term Warfarin therapy. The trial included 407 patients from 41 US centers. Enrolled patients experienced significantly less complications and a 95% implant success rate.

Dr. Ibrahim noted that one of the challenges with the PREVAIL trial, pertained to the efficacy endpoint. The first co-primary efficacy endpoint at 18 months follow-up was missed because of the exceptionally low number of strokes that occurred within the Warfarin group (control). The WATCHMAN™ device did achieve non-inferiority, however, for the second co-primary efficacy endpoint (stroke or systemic embolism >7 days after the randomization).

Regarding the question whether the WATCHMAN™ device is an appropriate alternative to Warfarin therapy Dr. Ibrahim noted that it is unsuitable since Warfarin is no longer the standard of care.

Next he discussed the Aspirin and Plavix^® registry (ASAP registry). The ASAP registry is a non-randomized feasibility study designed to determine if the WATCHMAN™ device is a safe and effective treatment for people unable to take Warfarin. The registry included 125 patients from 4 European centers. The average CHADS₂ score of enrolled patients was 2.7 and the post-procedure anti-platelet regimen included intake of clopidogrel over six months as well as an indefinite uptake of Aspirin. Patients were followed-up for up to one year after 3, 6, 12, 18, and 24 months. In addition, A TEE was conducted at 3 and 12 months to check proper placement of the device and closure of the LAA. The ASAP registry showed 3 ischemic strokes (3.75% vs 7.37% expected rate) and 4 patients with device-related thrombus.

Dr. Ibrahim also discussed the Multicenter Experience with the Amplatzer Cardiac Plug (ACP). He noted that the Amplatzer Cardiac Plug (ACP) is a well-proven technology thus far. The largest registry (969 patients and 1216 patient follow-up years) has recently been published wherein the three main indications for LAAO were: previous major bleeding, high bleeding risk, and coronary (CAD) stenting. The registry demonstrated a high success rate of 97% with a 4% rate of complications (0.6% death rate) and, in terms of efficacy, a 60% reduction of bleeding risk and a 60% reduction in stroke.

What are considered to be the best medical practices for anticoagulation or antiplatelet therapy for implantable endocardial LAA closure devices being studied, both pre-operatively and post-operatively?

Dr. Ibrahim noted that pre-operatively: most patients are treated with antiplatelet agents following bleeding episodes.

Post-operatively:

Following WATCHMAN: at least 6 weeks of anticoagulation followed by 6 months of DAPT before monotherapy.
Following Amplatz Cardiac Plug: 3 to 6 months of DAPT (based on the experience with other Amplatzers) followed by monotherapy.

Are there other available therapies (e.g. AF ablation) that are changing what an appropriate comparator group would look like for clinical studies?

Dr. Ibrahim mentioned two future randomized studies:

LAAO versus antiplatelet agents in patients with bleeding or at high risk of bleeding;
LAAO versus NOACs (Apixaban) in patients eligible to anticoagulation.

Dr. Ibrahim noted that AF ablation is not an alternative to anticoagulation.

Is surgical closure of the LAA still frequently done if there is a concomitant open heart procedure being done on a patient deemed to be at risk for stroke?

Dr. Ibrahim noted that LAA occlusion is still being performed during surgery for the following patients/procedures:

In patients with thrombus in the LAA;
During valve repair or replacement;
In the context of research projects.
In patients with bleeding on anticoagulation.

In conclusion Dr. Ibrahim noted that there is a clear clinical need for an alternative to oral anticoagulation in patients with AF. Left Atrial Appendage Occlusion is a safe and effective procedure to prevent ischemic and bleeding events and it should be considered as an alternative to Warfarin and/or for patients who bleed on anticoagulant therapy.

6. TAVI - Clinically Significant Technological Features

Dr. John Webb, Committee Member

There are several new unlicensed TAVI products with new design features that may make them superior for an aspect of the surgical procedure compared to licensed devices. Discuss the potential benefits of the new designs over licensed alternatives. Health Canada would like to better understand how much these new features may benefit the patient and when they become reason enough to allow access to an unlicensed device through the Special Access Program (SAP).

In your discussion, consider the following patient specific concerns and whether the newer designs address them.

Coronary ostium height;
Sinus of Valsalva diameter;
Degree of calcification;
Oblong or irregular shaped annulus;
Valve size;
Need for embolic protection;
Route of delivery.

Dr. Webb declared his affiliations and proceeded to provide a brief overview of transcatheter valves before addressing the discussion points outlined by Health Canada.

He noted that currently there is a wealth of data on transcatheter valves including data from the CoreValve US pivotal trial high risk study; a randomized, controlled trial comparing CoreValve transcatheter aortic valve implantation (TAVI) to surgical valve replacement in a high risk patient population. The trial data demonstrated a dramatically lower mortality rate in high risk (7.8% STS score) patients.

He explained that over the last decade, the number of TAVI patients has increased, especially as vascular complications and death rates decrease. The draw to TAVI is largely being driven by morbidity; TAVI patients have shorter hospital stays, less post-procedural pain and a quicker recovery in terms of mobility. However, as the number of patients accepted for TAVI increases, waitlists grow longer and, consequently, more patients die before they receive treatment.

Transfemoral (TF) TAVR reduces procedure time and has fewer complications due to the smaller catheter sizes now available. These favorable outcomes have led to increased funding.

He described some of the New Valve Features such as:

Repositionable valves: the key to a successful implantation is control and accuracy. Valves can be implanted too high or too low which may cause leaks and embolisms. In order to address the concerns over implantation, repositionable valves were developed to allow for valve repositioning after deployment.
Self-positioning valves: motorized delivery allows for stable deployment and single operator use. Self-positioning valves are designed to facilitate multiple interventional and surgical delivery approaches.
Valves with clip delivery: these valves actively clip onto the native leaflets which allow implantation of calcified valves.
Bicuspid valves have been considered an exclusion for transfemoral TAVI, however, with new features being introduced, newer valves do very well in the bicuspid.

Different valves have different implications (e.g., a self-expanding valve may not have good hemodynamics if it has not been fully expanded) and these differences influence the decision to choose one valve over another for any particular patient.

Stroke and TAVI

The risk of stroke is lower with TAVI than with surgical approaches.
The Claret Medical Sentinel™ is a popular embolic control device designed to capture and remove debris dislodged during endovascular procedures. It utilizes a proximal embolic filter delivered to the brachiocephalic artery, and a distal embolic filter delivered to the left common carotid artery. A large randomized trial using this device is ongoing.

Paravalvular Leaks and TAVI

A paravalvular leak is a complication associated with the implantation of a prosthetic heart valve whether surgical or a TAVI approach. A leak occurs when there is a lack of appropriate sealing between the structure of the implanted valve and the cardiac tissue and even mild leaks can be associated with late mortality. Improper sealing can occur if:
- The valve is too small;
- The valve is deployed too high;
- The valve is deployed too low;
- The valve is under-expanded.
Possible Solutions
- Self-expanding valves: while these valves may address the issue of fit, they tend to recoil and repositioning techniques such as snare will cause a greater risk of stroke.
- Variety in valve sizes: there is now a larger variety of valve sizes and physicians have become better at sizing. The gaps between each valve size, however, remain significant.

Coronary Occlusion

Coronary occlusion can be a life-threatening complication of TAVI. A patient being considered for TAVI is considered high risk if their coronary artery is less than 12 mm from the annulus because the risk of the native leaf pushing up over the coronary artery and causing a coronary obstruction is higher.

Native Leaflets Incorporated: JenaValve, Engager, Helio Sapien

The JenaValve™ TAVI System includes additional features to first generation valve implantation systems. It is repositionable during implantation procedure to ensure correct and precise positioning within the native valve.
The Engager seals the annulus by capturing the native leaflets with the control arms and conforming to the anatomy with the self-expanding frame. This feature helps reduce the risk of paravalvular leaks and coronary obstruction.
The Helio dock system acts as an anchor to help stabilise the SAPIEN XT valve for patients with aortic insufficiency. The native leaflets in the heart are captured between the transcatheter heart valve and the dock.

Durability

Taking all TAVI studies into account, the mean age of TAVI patients is 83 years old. In an 83 year old patient, the TAVI valves are expected to last for 5.5 years. Routinely, TAVI valves will last for at least 5 years, however, many will last longer.

Dr. Webb noted that more surgeons are moving towards bio-prosthetic valves. The durability associated with valve-in-valve implants is unknown; however, they are expected to have a shorter lifespan since they are not fully inflated. It should be noted that some surgical valves are not suitable for valve-in-valve procedures. For example, with a small surgical valve, a valve-in-valve procedure would not be suitable simply due to the hemodynamics.

Dr. Webb shared his own experience with the stentless valve with the Committee. He mentioned that, in his experience, these valves did not last as long because they did not possess the structural support. The stentless valves began to fail between 7 and 9 years after the procedure was done.

Dr. Webb stated that the issue of durability, is only an issue for intermediate risk patients (with an average age of 60) since they will likely outlive their valve. For these intermediate risk patients, however, the valve-in-valve procedure becomes an option.

The “Perfect” Valve

Dr. Webb presented a list of features that would contribute to a “perfect” valve:

Lower profiles to minimize vascular injury;
Femoral, apical, aortic, and subclavian access options;
Features that facilitate accurate positioning;
Repositionable/removable abilities;
Zero risk of paravalvular leak (e.g., have various types of sealing)

Pacemakers

The cost associated with pacemakers is higher because of the following factors:

Longer hospital stays;
Long-term monitoring;
They need to be replaced.

For the above-mentioned reasons, pacemaker and core-valve referral rates have fallen to between 12 and 30%.

Health Canada is becoming increasingly aware of some imaging abnormalities associated with some of the TAVR devices, specifically with leaflet motion. Could you please comment on that?

Dr. Webb spoke about a US trial in which, on occasion, the CT scan would show that one of the leaflets did not open. None of the patients in the trial experienced any clinical events. Therefore, he noted that one can stipulate that what’s happening here is related to an imaging abnormality rather than to sub-clinical outcomes.

Dr. Webb, Dr. Ibrahim and Dr. Yee left the meeting. Dr. Yee will return for his presentation.

7. Aortic Dissection devices - Treatment approach, patient population, dissection type, appropriate control

Dr. Prasad Jetty, Guest Speaker &

Dr. Thomas Forbes, Guest Speaker

Dr. Prasad Jetty and Dr. Thomas Forbes were invited to address this topic. They introduced themselves and both indicated that they had no disclosures to declare and proceeded to address the topics identified by Health Canada. Dr. Jetty opened the discussion. He provided a brief overview of aortic dissection.

He noted that an aortic dissection is a spontaneous tear in the innermost layer of aorta (intima) allowing blood to be driven into the media layer, creating two channels for blood flow. If the tear continues, it can interfere with the visceral vessels or could potentially lead to a stroke. Aortic dissection is typically caused by hypertension and is often uncontrolled. It is a rare condition that affects men and is observed more often in Caucasians and African Americans.

He explained that there are two types of aortic dissections:

Uncomplicated: not interfering with any of the branch vessels.
Complicated: the false lumen can expand and rupture or compress the true lumen. If the carotid artery is affected, a stroke can occur.

Dr. Jetty noted that an aortic dissection is caused by a tear in the intima, whereas, an aneurysm is a gradual deterioration of the aortic wall. It is asymptomatic and risk of rupture and death are dependent on diameter of aneurysm.

Currently, the gold standard for identifying an aortic dissection is a CT scan. This scan will help identify the location of the tear which will allow for appropriate classification of the dissection. Classification determines how to proceed with management of the tear. If the tear is in the ascending aorta, for example, it can propagate and is considered a Type A dissection (surgical emergency, close to the coronary and arch vessels, risk of death is significant). If the tear occurs in the descending thoracic aorta, it is considered a Type B dissection. Management for a Type B dissection is determined by whether the dissection is considered complicated or uncomplicated.

Type A Dissection Repair:

Emergency cardiac surgery;
resection of intimal tear;
replacement of the affected aorta with graft;
reconstitution of flow through true lumen;
25% operative mortality.

Type B complicated (rupture or malperfusion) Dissection Repair: There are two options for repair:

Option one: an open surgical repair. For many years open surgery was the only option. However, due to the high mortality rates, this type of management has been abandoned for acute Type B aortic dissections.
Option two: an endovascular repair with a stent graft. Through small, percutaneous incisions, a device is delivered through the femoral artery into the thoracic artery and a stent is deployed. The goal of the repair is to seal the proximal tear site, re-expand the true lumen and obliterate the false lumen or have it thrombus. This type of management is less invasive and has a much lower morbidity and mortality than the surgical repair.

A question was posed to Dr. Jetty regarding the possibility for a randomized control trial that looked at endovascular repair using stent grafts in Type B dissections. Dr. Jetty noted that stent grafts have a good track record in the aorta for thoracic aneurysms. The incidence rate of type B dissection is extremely low; therefore, a randomized control trail would not be feasible or ethical.

STABLE Trial on Aortic Dissection was described. Rather than opening the chest cavity and clamping off the aorta to surgically implant a graft to treat the damaged section of the thoracic aorta, physicians insert a catheter loaded with a self-expanding, fabric covered stent-graft through a surgical opening in the femoral artery. The catheter is guided through the patient’s blood vessels under fluoroscopy until the device is positioned across the dissected section of the aorta. The stent-graft then expands upon deployment from the catheter to reopen the original path through the aorta and should reduce blood flow into the false pathway in the damaged vessel wall, thereby restoring normal aortic blood flow. The uncovered Zenith Dissection Stent is used to expand the true lumen in the distal thoracic aorta where preservation of the side branch artery blood flow is critical.

Type B uncomplicated Dissection Repair: the visceral vessels and the carotid are not compromised in an uncomplicated dissection. A significant proportion of uncomplicated dissections become complicated (in the range of 30%) and, after these patients’ aorta is dissected, they do not have the same life expectancy (after 5 years, approximately 60% have died). As is the case in a complicated repair, there are two options:

Option one: optimal medical therapy in the form of antihypertensive medication and monitoring.
Option two: intravascular stent repair.

Traditionally, in Type B uncomplicated dissection, patients are treated non-operatively. Patients will be put on optimal medical therapy and will require lifelong CT surveillance to ensure the false lumen does not expand and rupture.

Dr. Jetty concluded his presentation with a case report in which the STABLE procedure was used to address a Type B uncomplicated dissection repair.

Dr. Forbes continued the discussion. He outlined the objectives of his talk:

Incidence, etiology, classification, progression, outcome of aortic dissection;
Difference between Aortic Dissection (AD) and aneurysms.
Clinical guidelines for acute and chronic AD;
Evidence for treatment of AD with stent grafts;
Challenges for RCT to compare open surgery vs. stent graft vs. medical management;
Stent graft requirements for AD’s;
Evidence of necessary for PMA of new stent graft;
FDA approach, i.e., single arm studies etc.

Dr. Forbes reviewed the difference between the two types of dissections:

Type A: involves the ascending aorta.
Type B: involves the descending aorta.

Type B dissection involves the aorta from the top down to the thoracic and abdominal aorta. It can also involve the branches (to the spinal cord, legs, kidneys, etc.). Type B Aortic Dissections
Are defined as follows:

Acute: within 2 weeks of dissection (occasional patient with a silent dissection with no symptoms). Can be misdiagnosed as a myocardial infarction (MI).
Subacute: 2 weeks to 3-6 months. Longer term complications.
Chronic: greater than 3-6 months.

Acute uncomplicated: all patients with type B aortic dissection will undergo hypertension control. If the pain resolves and the aorta does not expand or rupture, these patients are denoted as acute uncomplicated.

Acute complicated: within the first 2 weeks of dissection but the patient is failing medical management. The complications can include: aorta expansion, persistent pain, or the new channel can compress the blood supply to important organs called a malperfusion.

Chronic: if the patient survives the first acute phase, they are left with a dissection and will have a lifelong dissected aorta which may or may not dilate depending on progression of medical management. If the dissection becomes complicated, the patient will develop an aneurysm.

Acute Aortic Syndrome

There are a number of pathologies that can present in the thoracic aorta; these include aortic dissection, intramural thrombus, and penetrating atherosclerotic aortic ulcer.

Dr. Forbs described available Clinical Guidelines:

The Canadian Guideline “CCS/CSCS/CSVS Position Statement on Thoracic Aortic Disease Interventions” is currently being written.

Dr. Forbes noted that the US Clinical Guidelines were published in 2010. He highlighted that the Guidelines recommend that acute dissections be managed medically unless they are complicated. When the dissection is complicated, the recommendation for management was endovascular repair.

He summarized recommendations of the European Guidelines:

For Type A: surgery.
For Uncomplicated Type B: medical therapy.
For Complicated Type B: endovascular repair.

He noted that prior to Stent Grafts:

Uncomplicated Type B aortic dissections: treated medically.
Complicated acute Type B AD’s: treated with open surgery, which was associated with high mortality, morbidity and poor long term outcomes.
Currently, Rx of complicated acute and aneurysmal chronic Type B AD’s has shifted towards stent grafts.

Dr. Forbes commented on three key aortic dissection trials:

Acute, complicated Type B AD—STABLE Trial.
Acute, uncomplicated Type B AD—ADSORB Trial.
Sub-acute, uncomplicated Type B AD—INSTEAD-XL Trial.

The STABLE Trial focused on complicated dissections. It was a small study and most of the patients were acute (earlier in presentation, the higher the complication rate). Also, the majority of the patients had a degree of dilation and malperfusion. An uncovered dissection stent, specifically designed for this indication was used. The aim of the trial was to expand the stent so it could be placed across the vessels. If a covered stent had been used, the origin of the vessels would be occluded.

The ADSORB Trial focused on acute, uncomplicated dissections. The trial compared optimal medical therapy (OMT) to TEVAR and OMT. In this trial, rather than patient-specific outcomes, the main endpoint was aortic remodeling. There were 30 patients randomized to each group (no in-hospital deaths). Initial results for the trial showed that the TEVAR and OMT group had better results in terms of aortic remodeling. The long term data has not yet been obtained.

The INSTEAD-XL Trial looked at endovascular repair of Type B aortic dissection. It published longer term data of the randomized investigation of stent grafts in aortic dissection. The two year data showed no advantage to a stent graft while the 5 year data showed improved survival, fewer deaths due to aortic problems and non-aortic problems.

Dr. Forbes outlined the main predictors of aortic dissection as follows:

Younger age;
Larger aortic diameter;
A large false lumen that does not thrombose in the acute or sub-acute phase;
Degree of false lumen thrombosis;
The number of tears between the two channels.

He provided a summary of existing evidence:

Acute complicated type B dissections, whether aortic expansion or malperfusion, are treated endovascularly with stent grafts.
If possible, a period of medical management might be beneficial since results in the subacute phase are better than in the acute phase.
Some long term benefit to patients treated with stent grafts in the subacute phase (INSTEAD-XL).
It remains unclear whether acute, uncomplicated Type B aortic dissections should be treated early (with a stent graft) or not (ADSROB).

He noted that dissections are very complex and require lifelong management. Endovsacular repair of Type A aortic dissections has some status of approval on a compassionate level for patients that are turned down for open repair.

Stent Grafts for Type B Aortic Dissections

Dr. Forbs noted that most stent graft systems are designed for aneurysms. There are covered and uncovered stents.

Some of the different features include:

Proximal configuration—retrograde dissection (bare metal upper stent to allow secure fixation). Some stents have hooks or barbs that latch on to the aorta to hold it in place and others are covered right proximal.
A retrograde Type A dissection usually results in death. Some predictors are design features such as a bare stent. Barbs have been associated with retrograde Type A dissections; therefore, many surgeons chose their grafts for dissections differently than they chose them for aneurysms. It could be more of an oversizing issue than a proximal component of the stent graft since, when measuring the diameter of the aorta; the need to oversize is even less than in other procedures.

Next he provided an overview of the FDA Regulatory Approach:

TEVAR for Type B aortic dissections and BTAI had become “standard of care” in US, but no approved TEVAR device for these indications.
Gore TAG device approved for TAA but not for other pathologies.
Optimally RCT data would support new indications but low incidence, variability, lack of clinical equipoise between stent graft and open repairs.
Single arm studies.
With absence of control arm, need a prospectively defined measure of success.
For dissections, data was collected from five IDE study, a common set of definitions was set and applied retroactively.
Master file provides comparative group.
Pre and Post-market studies, non-pathology specific approval.

FDA Regulatory Response

FDA met with societies and industry to best evaluate stent grafts to support labels that would cover all the pathologies they were used for after their approval for TAA in 2005.
Because of difficulty with studies (RCT etc.), FDA agreed to allow a broad indication for an individual endograft (rather than specific lesion type).
Submission of safety data for treatment of TAA, BTAI, acute complicated Type B aortic dissections.

This concluded the presentation.

Dr.’s Jetty and Forbes thanked the Committee for the opportunity to present at this meeting and left the meeting.

Dr. Menkis left the meeting.
Dr. Yee returned to the meeting to deliver his presentation.

8. Intracompatibility between IS-4 Leads and Cans

Dr. Raymond Yee, Committee Member

In advance of the release of ISO 27186 - Four pole connector system for implantable cardiac rhythm management devices - Dimensional and test requirements the Scientific Advisory Committee was asked to comment on appropriate clinical study requirements. Since then, both IS4 and DF4 leads, as well as compatible pulse generators have been licensed in Canada.

Dr. Yee declared his affiliations and proceeded with a brief introduction before addressing the questions posed by Health Canada.

Dr. Yee noted that there were two basic standards that dealt with the connection between the lead and the generator. The first standard is the IS-1 standard (introduced in early 1990s) which governs unipolar and bi-polar, low voltage pacing leads. There was then a DF-1 standard which covered unipolar defibrillation coils. Since there are now leads that require more than 2 electrodes, there was a need for new standards. The DF4/IS4 standards apply to leads with 3 to 4 electrodes.

He explained that the IS4 is a low voltage pacing lead and the DF4 is a high voltage lead involving pacing and defibrillation.

He discussed the adoption rate of DF4 Leads vs. DF1 Leads. Initially, the percentage of DF4 leads was less than 10%, however, that number has now settled in at about 20-25%. The vast majority of leads implanted now are DF4 standard. Other manufacturers show a similar trend. This trend suggests that the DF4 has been adopted by the medical community.

He summarized issues addressed by IS4/DF4:

Interchangeability between manufacturers;
Lead complexity;
Operator errors - misconnects;
- Reduced the complexity. With only one pin, there is less risk of operator error.
Hardware bulk.
- The yoke is immediately removed which was a source of physician complaint.

He explained that when comparing a DF1 and a DF4, the DF4 is much more complex. With the DF4, all four electrodes form one connector and go into one port which streamlines the port.

Dr. Yee described issues Created by IS4/DF4:

Legacy IS1/DF1 systems and generator replacements;
High implant DFT’s (Cogert, HRJ 2012);
Isolated component failures;
Pacing
- If the physician’s preference is to implant the lead with the tip at the apex, another lead must be added for pacing. This cannot be done with the DF4.
- If the pacing sensing component of a lead fails, the DF4 lead must be abandoned and replaced with a DF1 lead.
Sealing ring/connector variability (Bracke, HRJ 2014).

After this brief overview Dr. Yee proceeded to address the questions posed by Health Canada.

Provide an update on the available clinical data for the DF4 leads, especially as it relates to adverse events, lead fractures, inconsistent results between lead manufactures, and anything which should be given special consideration during regulatory review.

Dr. Yee noted that there is no major evidence suggesting that there is a signal of concern for reliability in DF4 leads or IS4 leads. There has only been one IS4 issue reported where there was an insulation issue in which the conductor wires on the inside could abrade and permeate through the insulation of the lead.

Dr. Yee mentioned that there are no concerns of integrity of the DF4 family. It has been 2-4 years now and the medical community has not observed any signal suggesting reliability problems.

Provide an update on the available clinical data for the IS4 leads, especially as it relates to rates of phrenic nerve stimulation, adverse events, and claims of improved CRT response, and anything which should be given special consideration during regulatory review.

IS4 LV leads (quadripolar LV leads): limited publication on some of the manufacturers.

Dislodgment rates are not zero (the hope for the IS4 leads was that the likelihood of stability would increase so that the two best placed poles are chosen). Also, there is no control over the caliber of the veins.

Regarding nerve stimulation, there have been four reports where the rates of femoral stimulation range from 10% to 40%, vein choose is crucial. If a lateral vein is chosen, nerve stimulation can be avoided. Having four electrodes grants more flexibility and can eliminate stimulation issues (though not entirely).

Clinical Response

There is no clinical response data currently available. There is acute and short term data dealing with dual LV site pacing. A clinical trial will be starting in Canada to look at the use of quadripolar and clinical response to CRT therapy.

Currently, two kinds of data exist:

Data looking at dual LV site pacing where two separate leads are placed into two separate veins (different from quadripolar leads).
Three small sample size studies comparing routine pacing versus dual LV site pacing in two separate veins.
- The first study used echo and results demonstrated that better echo parameters (time for ejection of blood) were reported suggesting the dual pacing was slightly better.
- Similarly, the TRUST study claimed that NYHA improved by 1.4 classes versus 1 class.
- The third study suggested that patients could walk further (25 meters).

Dual site LV pacing has two separate leads in two separate veins. The IS4 quadripolar LV leads do not have that, they put all four electrodes in one vein and there is very little data to support that that works.

There is no long term data showing that response rate to CRT using the ISF improves response.

The interchangeability between implantable leads and pulse generators from different manufacturers was to be one of the benefits of ISO 27186, though there are reports of failed interoperability, is this a concern amongst the clinical community in Canada?

Dr. Yee highlighted one area, noting that it may not be the root cause of the problem. The dimensions and properties of the leads are well specified. The difference between the IS4 and IS1 are the sealing rings. On the IS1, the sealing rings are reasonable for blocking fluid from ingressing into the port (causing a short out of the connection). With the DF4 standard, there are no sealing rings on the lead; they are on the connector blocks themselves. There is no way of taking a DF1 lead and pushing it into the DF4 block.

Dr. Yee stated that he had contacted companies to ask about what they thought was the root of the problem. The response he received was that the standard is fairly robust and specific but that there is one aspect that is not spelled out which is robustness. They noted that when it comes to pushing this lead into a DF4 connector head, user is informed how much force to apply in order to get it inserted properly, however, it is not specified what happens to the lead as it is being gripped and pushed. The segment just beyond what is covered by the DF4 standard may be more malleable than desirable which means that when the lead is pushed, it flexes.

DF4/IF4 incompatibility amongst manufacturers is an annoyance for physicians; however, it is not a major concern. The percentage of implants that mix manufacturers is unknown; however, it is likely the minority and can be overcome using mineral oil.

9. Late Breaking Issues in Cardiovascular Devices

Prior to moving on to closing remarks, the Committee participated in a brief discussion on late breaking issues on cardiovascular devices.

Bioabsorbable Vascular Scaffolds and their Appropriate Patient Populations
Ian Aldous , Medical Devices Bureau (MDB),

Round-table, Committee Members

Health Canada provided a brief presentation of the topic. It was noted that the bioabsorbable scaffolds present a unique platform for vascular stents. Some recent studies (ABSORB II) have suggested that drug eluting bioabsorbable scaffolds perform in a similar manner to some of the other traditional drug eluting stents.

Currently, there are two types of bioabsorbable vascular scaffolds:
- PLLA stents: the stent is covered in a bioabsorbable polymer.
- An entirely bioabsorbable scaffold.
Health Canada posed number of questions to the committee members:

Are there any clinically important differences in terms of typical clinical use or clinical outcomes between the traditional DES and drug eluting bioabsorbable scaffolds?

Is there a particular patient population for which drug eluting bioabsorbable scaffolds are better suited than traditional stenting? Who do you predict will get bioabsorbable scaffolds in Canada if one were to be licensed?

The Committee agreed that if there is a licensed alternative on the market, the new product needs to show superiority. Currently we have a gold standard: flexible, drug eluting, second generation stents.

Regarding a particular patient population, the Committee noted that patients in need of re-stenting could be good candidates since these stents are being sold as potentially less thrombogenic and better for “access to site” for future procedures.

The risk profile of bioabasorbable stents is different than traditional stents in that they are not permanent. How important is it to you as a clinician that the stent is not a permanent implant? Based on the different risk profile of a bioabsorbable scaffold, to what should they be compared to? (i.e., DES, BMS, angioplasty, medical therapy, non-inferior vs. superior).

The Committee noted that there is a theoretical disadvantage because the structure disappears after 12 months; the vessel may have contracture that is prevented by the presence of the scaffolding.

The main difference is that bioabsorbable stents are bulkier, hard to introduce. Thrombosis rate is not better. There does not seem to be any advantage in using bioabsorbable stents at this time.

The Committee recommended that the device must be equivalent to the gold standard stent, especially with regard to thrombosis and restenosis.
Dual Antiplatelet Therapy- Responses to dual antiplatelet therapy duration clinical trial results from AHA

Kevin Day, Medical Devices Bureau (MDB),
Round-table, Committee Members

Late-stent thrombosis concerns erupted in 2006, have triggered a flurry of investigation on how long dual antiplatelet therapy would need to be continued. The many observational and predominantly small- to moderate-sized studies were inconsistent and inconclusive, though, leading the FDA to mandate what became the DAPT trial. It randomized people to discontinue use of clopidogrel (Plavix) or prasugrel (Effient) plus aspirin either 12 or 30 months after coronary artery stenting. The risks and benefits of long-term dual antiplatelet therapy remain unclear. Optimal duration of dual antiplatelet therapy (DAPT) after DES implantation might be shorter than currently recommended 1-year interval.

The Committee was asked to comment on optimal duration of dual-antiplatelet therapy. Members stated that there are two reasons for which a patient is put on antiplatelet therapy after a stent is implanted. The first is due to the stent and the second is due to the other arteries. If a patient is put on dual-antiplatelet therapy for 1 month versus 12 months, there is a 20% reduction in myocardial infarction and stroke at 12 months. It is not surprising, noted the Committee, that there may be some benefit to prolonged dual antiplatelet therapy like less stent thrombosis and less thrombosis in other arteries.

10. Next Steps, Closing Remarks and Adjournment of Meeting